Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

Most AI memory systems stop at retrieval. Store information. Search information. Return information. EverOS 1.0.0 is exploring something much more interesting: What happens after memory is retrieved?🧵

20,757 görüntüleme • 27 gün önce •via X (Twitter)

0 Yorum

Yorum bulunmuyor

Orijinal gönderinin yorumları burada görünecek

Benzer Videolar

🚨 SCIENTISTS SAY “MAGIC” MAY BE WHAT GIVES SPACE-TIME ITS GRAVITY. For years, physicists have understood how entanglement can build the structure of space-time in holographic models. But something was missing: why does space-time curve in response to matter the essence of gravity? A team including Charles Cao and John Preskill now proposes the missing ingredient is a quantum property called “magic” a measure of how complex and non-classical a quantum state is (the kind that makes quantum computers hard to simulate classically). In their theoretical framework, adding this magic turns rigid space into something that can bend. Matter can now tell space how to curve. Why this matters: • It offers a new way to think about how gravity emerges from quantum information • It connects ideas from quantum computing (error correction, magic states) directly to fundamental physics • It suggests space-time itself may be one of the most quantum objects in existence The deeper implication: Gravity may not be a fundamental force at all. It may be what happens when quantum information becomes sufficiently complex and “magical.” This is still early theoretical work in specific holographic models. But it hints that the pliability of the universe might have quantum roots we are only beginning to understand. What do you think is gravity ultimately just extremely complicated quantum information, or do you think we’re still missing something much deeper? Follow for more frontier quantum gravity and quantum information research.

TheNewPhysics

15,329 görüntüleme • 1 ay önce

There’s been two papers released in the past couple months, one by Google and one by NVIDIA, that argue that ordering the documents retrieved by RAG systems can enhance performance. However, they both give two different strategies on HOW these documents should be ordered 🤔 Both papers agree on two main points: 1️⃣ There’s a fundamental issue in RAG - as more documents are retrieved, more irrelevant context (e.g., hard negatives) are introduced, which leads to confusion for the LLM and eventually degrades the quality of the generated output. This is called an inverted-U performance curve. 2️⃣ Ordering the retrieved documents is a key lever for optimizing RAG performance. Google Cloud researchers proposed ordering results based on relevance scores: The authors in this paper argue for relevance-based reordering, or ordering the retrieved chunks based on their similarity scores, so the most relevant documents are at the beginning and the end of the inputs to counter the “lost in the middle” effect. NVIDIA researchers proposed ordering results based on the original sequence of document chunks: The authors of this paper argue for Order-Preserving Reordering, or Order-Preserve RAG (OP-RAG), to maintain the logically coherent content flow of the document. So they preserved the original order of retrieved document chunks in the source text, instead of ranking them by relevance scores. So which one is right? It probably depends on the specific use case and dataset - relevance-based reordering could perform better in tasks where you need fast access to the most critical information (e.g., fact retrieval, QA systems), while order-preserving RAG might be better where you need to understand the sequential structure of information (e.g., narrative or legal documents). There are still so many uncertainties in AI - we don’t actually know what we’re doing, and it takes awhile to figure out the best strategies for most things! Excited to see more research about this.

Victoria Slocum

15,333 görüntüleme • 1 yıl önce